TEST #1786 RESULTS
11/12/2025 - 2:05 PM
ACCELERATOR
MODEL
18.8
tokens/s
generation
1.20
sec
time to first token
1138
tokens/s
prompt
261
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.6GB
OS
Linux
Kernel Release
6.11.0-1016-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #16-Ubuntu SMP PREEMPT_DYNAMIC Sun Sep 21 16:52:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1409
tokens/s
19.1
tokens/s
781
ms
pp4096+tg256
994
tokens/s
17.1
tokens/s
4.18
sec
pp2048+tg256
1244
tokens/s
18.4
tokens/s
1.70
sec
pp2048+tg768
1251
tokens/s
18.3
tokens/s
1.69
sec
pp1024+tg1024
1413
tokens/s
19.1
tokens/s
782
ms
pp1280+tg3072
1355
tokens/s
17.8
tokens/s
1000
ms
pp384+tg1152
1552
tokens/s
19.5
tokens/s
301
ms
pp64+tg1024
866
tokens/s
19.7
tokens/s
183
ms
pp16+tg1536
162
tokens/s
19.8
tokens/s
150
ms